Patent Application
20240101604
This disclosure relates to peptides that bind MENA and methods of using such peptides in the treatment of cancer.
Cytoskeleton-regulating Ena/VASP proteins are recruited to different regions of the cell by binding proline-rich short, linear motifs (SLiMs) via their N-terminal EVH1 domains and promote actin polymerization via their C-terminal EVH2 domains. The three paralogous family members in humans, ENAH Also called MENA or hMena), EVL, and VASP, are implicated in many cellular functions. For example, recruitment of Ena/VASP paralogs by FYB1 to phagocytic cups is essential for phagocytosis in macrophages. Ena/VASP proteins are also important in regulating cell motility, and in axon guidance through the Netrin-1/DCC pathway. Dysregulation of Ena/VASP proteins results in pathologies including cancer metastasis and arteriosclerosis establishing them as attractive therapeutic targets. In particular, ENAH and its isoform ENAHINV are highly expressed in invasive cancer cells, and these proteins have been proposed both as biomarkers and as candidate therapeutic targets for metastatic disease. Previous work has explored the development of ENAH-binding mini-proteins and small-molecule inhibitors. The Ena/VASP EVH1 domain recognizes the SLiM [FWYL]PXΦP, where X is any amino acid, and Φ is any hydrophobic residue. This motif (“FP4”), readily adopts a polyproline type II (PPII) helix structure and binds weakly to the EVH1 domain. Searching for the low complexity FP4 motif in the human proteome, using the definition above, yields 5067 instances. Not all of these motifs have the opportunity to engage Ena/VASP proteins in the cell, given that spatial, structural, and temporal context are important determinants for cellular interaction. But the number of potential interaction partners is clearly large and raises the question of whether sequence determinants beyond the FP4 motif affect molecular recognition specificity.
Ena/VASP EVH1 domains are part of a larger EVH1 family also found in WASP, SPRED, and Homer proteins, many of which converge to regulate the actin cytoskeleton. Confoundingly, the binding specificity profiles these distinct EVH1 domains overlap, when considering only their low complexity SLiM profiles. The extent to which these domains cross-react vs. bind selectively in the cell is unknown. Even within the Ena/VASP family, there are outstanding questions about EVH1 interaction specificity. ENAH, VASP, and EVL proteins have distinct functions, subcellular localizations, and binding partners. For example, knockdown of ENAH or VASP in migratory breast cancer cells decreases their invasive potential, while EVL is a known suppressor of breast cancer cell invasion. Yet the three proteins are 100% identical in sequence in the core FP4 binding groove and share 62-72% sequence similarity throughout the EVH1 domain. This creates considerable challenges to creating selective peptides useful as therapies.
There is increasing evidence that Mena and isoform of Mena (MenaINV) may be useful therapeutic targets and biomarkers (Oudin et al. (2017) Trends in Cancer 3:7). Evidence from a murine model suggests that Mena deficiency delays tumor progression (Roussos et al. Breast Cancer Research 2010, 12:R101) Moreover, Mena confers resistance to paclitaxel in certain triple-negative breast cancers (Oudin et al. (2011) Mol Cancer Ther 16:143).
Described herein is a method for treating cancer comprising administering a composition comprising a polypeptide comprising the amino acid sequence LPPPPMEVLMDKSFASLES. In various cases, the polypeptide comprises an amino acid sequence selected from the groups consisting of: NLEHLPPPPMEVLMDKSFASLES, EMEGNLEHLPPPPMEVLMDKSFASLES, and NLEHLPPPPMEVLMDKSFASLESGSGSPPPPPF. Also described is a method for treating cancer comprising administering a composition comprising a polypeptide comprising the amino acid sequence LPPPPMEVLMDKSFASLES with no more than three single amino acid substitutions and a method for treating cancer comprising administering a composition comprising a polypeptide comprising the amino acid sequence FDDFPPPPPPPPVDYEDLPSISGNFPPPPPL or a polypeptide comprising the amino acid sequence FDDFPPPPPPPPVDYEDLPSISGNFPPPPPL with no more than 2 single amino acid substitutions. In various cases, the polypeptide is 16-60 or 23-60 amino acids long. In various cases the cancer is selected from the group consisting of: breast cancer, cervical cancer, colorectal cancer, and pancreatic cancer.
Also described is: an isolated polypeptide comprising or consisting of the amino acid sequence LPPPPMEVLMDKSFASLES,
FDDFPPPPPPPPVDYEDLPSISGNFPPPPPL as well as
a pharmaceutical composition comprising any of the isolated polypeptide and a compound comprising any of the isolated polypeptides covalently linked to a moiety that facilitates penetration of a mammalian cell.
In various cases, the method further comprising administering a taxane chemotherapy (e.g., a taxane chemotherapy is selected from the group consisting of paclitaxel, albumin-bound paclitaxel and docetaxel). Also described is a method for reducing resistance to a taxane chemotherapy in a patient suffering from a cancer (e.g., breast cancer), the method comprising administering a polypeptide described herein.
Also described is a method for treating cancer, the method comprising administering to a patient in need thereof, a nucleic acid molecule (e.g., an mRNA) encoding a polypeptide described herein.
In some cases of the compounds described herein, the peptide includes at least 16 amino acids. In some cases the peptide includes up to 60 amino acids (e.g., 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, etc.). In some cases the peptide no more than 60 amino acids (e.g., 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, etc.). In some cases, the compounds include at least one amino acid that is not one of the 20 common, naturally-occurring amino acids.
In some cases, the compounds described herein also comprise a detectable label. In some cases, the detectable label is linked to the peptide.
In some cases, the compounds described herein also comprise a moiety linked to the peptide. In some cases, the moiety can be a functional peptide, polyethylene glycol (PEG), alkyl groups (e.g., C1-C20 straight or branched alkyl groups), fatty acid radicals, and combinations thereof. In some cases, the peptide is linked to PEG.
In some cases of the compounds described herein, the peptide is linked to a second peptide or functional moiety. In some cases the second peptide or functional moiety modulates the activity of the compound. For example, in some cases the second peptide or functional moiety modulates the solubility of the compound, modulates the stability of the compound, modulates the ability of the compound to permeabilize a cell, acts to target the compound within/to the cell, labels the compound, modulates the affinity of the compound for MENA. Modulating the activity of the compounds described herein can be increasing or decreasing the activity.
In some cases of the compounds described herein, the peptides are modified. In some cases, the modification is selected from the group consisting of acetylation, amidation, biotinylation, cinnamoylation, farnesylation, fluoresceination, formylation, myristoylation, palmitoylation, phosphorylation, stearoylation, succinylation, sulfurylation, and combinations thereof.
In some cases of the compounds described herein, the peptides include at least one peptide bond that is replaced by a non-natural peptide bond. In some cases, the peptide bond is replaced by a bond selected from the group consisting of a retro-inverso bonds (C(O)—NH); a reduced amide bond (NH—CH2); a thiomethylene bond (S—CH2 or CH2-S); an oxomethylene bond (0-CH2 or CH2-0); an ethylene bond (CH2-CH2); a thioamide bond (C(S)—NH); a trans-olefin bond (CH═CH); a fluoro substituted trans-olefin bond (CF═CH); a ketomethylene bond (C(O)—CHR) or CHR—C(O) wherein R is H or CH3; and a fluoro-ketomethylene bond (C(O)—CFR or CFR—C(O) wherein R is H or F or CH3.
In some cases, the compounds described herein also comprise a carrier.
In some aspects, the present disclosure provides a pharmaceutical composition comprising the compounds described herein. In some aspects, the present disclosure provides a method of treating cancer comprising administering the compound described herein.
Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Methods and materials are described herein for use in the present invention; other, suitable methods and materials known in the art can also be used. The materials, methods, and examples are illustrative only and not intended to be limiting. All publications, patent applications, patents, sequences, database entries, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control.
Other features and advantages of the invention will be apparent from the following detailed description and figures, and from the claims.
Described below are studies that identified a 36-residue human proteome-derived peptide binders of the ENAH EVH1 domain. The studies identified three distinct ways in which local and distal sequence elements surrounding the FP4 motif modulate ENAH binding affinity and specificity. Also identified a network of dispersed residues in the ENAH EVH1 domain that allows it to adopt a conformation inaccessible to EVL and VASP, allowing ENAH to selectively bind to retinal protein PCARE. In the studies described below, this interaction were validated in cells. The studies below also describe a nanomolar EVH1 binder that is 400-600-fold selective for ENAH over VASP and EVL.
Amino acids are the building blocks of the peptides herein. The term “amino acid” refers to a molecule containing both an amino group and a carboxyl group as well as a side chain. Amino acids suitable for inclusion in the peptides disclosed herein include, without limitation, natural alpha-amino acids such as D- and L-isomers of the 20 common naturally-occurring alpha-amino acids found in peptides (e.g., Ala (A), Arg (R), Asn (N), Cys (C), Asp (D), Gln (Q), Glu (E), Gly (G), His (H), Ile (I), leu (L), Lys (K), Met (M), Phe (F), Pro (P), Ser (S), Thr (T), Trp (W), Tyr (Y), and Val (V), uncommon alpha-amino acids (including, but not limited to α,α-disubstituted and N-alkylated amino acids), common naturally-occurring beta-amino acids (e.g., beta-alanine), and uncommon beta-amino acids. Amino acids used in the construction of peptides of the present invention can be prepared by organic synthesis, or obtained by other routes, such as, for example, degradation of or isolation from a natural source.
There are many known amino acids beyond the 20 common naturally-occurring amino acids, any of which may be included in the peptides of the present invention. Some examples of uncommon amino acids are 4-hydroxyproline, α-4-pentenyl alanine, aminoisobutyric acid (Aib), cyclohexyl alanine (Cha), norleucine, desmosine, gamma-aminobutyric acid, beta-cyanoalanine, norvaline, 4-(E)-butenyl-4(R)-methyl-N-methyl-L-threonine, N-methyl-L-leucine, 1-amino-cyclopropanecarboxylic acid, 1-amino-2-phenyl-cyclopropanecarboxylic acid, 1-amino-cyclobutanecarboxylic acid, 4-amino-cyclopentenecarboxylic acid, 3-amino-cyclohexanecarboxylic acid, 4-piperidylacetic acid, 4-amino-1-methylpyrrole-2-carboxylic acid, 2,4-diaminobutyric acid, 2,3-diaminopropionic acid, 2,4-diaminobutyric acid, 2-aminoheptanedioic acid, 4-(aminomethyl)benzoic acid, 4-aminobenzoic acid, ortho-, meta- and/para-substituted phenylalanines (e.g., substituted with —C(═O)C6H5; —CF3; —CN; -halo; —NO2; CH3), disubstituted phenylalanines, substituted tyrosines (e.g., further substituted with -Q=O)C6H5; —CF3; —CN; -halo; —NO2; CH3), and statine. Additionally, amino acids can be derivatized to include amino acid residues that are hydroxylated, phosphorylated, sulfonated, acylated, and glycosylated, to name a few.
In some instances, peptides include only common, naturally-occurring amino acids, although uncommon amino acids (i.e., compounds that do not commonly occur in nature but that can be incorporated into a polypeptide chain) and/or amino acid analogs as are known in the art may alternatively be employed. Also, one or more of the amino acids in a peptide or polypeptide may be modified, for example, by the addition of a chemical entity such as a carbohydrate group, a hydroxyl group, a phosphate group, a farnesyl group, an isofarnesyl group, a fatty acid group, a linker for conjugation, functionalization, or other modification, etc.
In some instances, peptides can include 1, 2, 3, 4, or 5 single amino acid substitutions, e.g., conservative substitutions. In some instances, a “conservative amino acid substitution” can include substitutions in which one amino acid residue is replaced with another naturally-occurring amino acid residue having a similar side chain. Families of amino acid residues having similar side chains have been defined in the art. These families include amino acids with basic side chains (e.g., lysine, arginine, histidine), acidic side chains (e.g., aspartic acid, glutamic acid), uncharged polar side chains (e.g., glycine, asparagine, glutamine, serine, threonine, tyrosine, cysteine), nonpolar side chains (e.g., alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan), beta-branched side chains (e.g., threonine, valine, isoleucine) and aromatic side chains (e.g., tyrosine, phenylalanine, tryptophan, histidine).
The addition of polyethelene glycol (PEG) molecules can improve the pharmacokinetic and pharmacodynamic properties of the polypeptide. For example, PEGylation can reduce renal clearance and can result in a more stable plasma concentration. PEG is a water soluble polymer and can be represented as linked to the polypeptide as formula: XO—(CH2CH2O)n—CH2CH2—Y where n is 2 to 10,000 and X is H or a terminal modification, e.g., a C1-4 alkyl; and Y is an amide, carbamate or urea linkage to an amine group (including but not limited to, the epsilon amine of lysine or the N-terminus) of the polypeptide. Y may also be a maleimide linkage to a thiol group (including but not limited to, the thiol group of cysteine). Other methods for linking PEG to a polypeptide, directly or indirectly, are known to those of ordinary skill in the art. The PEG can be linear or branched. Various forms of PEG including various functionalized derivatives are commercially available.
PEG having degradable linkages in the backbone can be used. For example, PEG can be prepared with ester linkages that are subject to hydrolysis. Conjugates having degradable PEG linkages are described in WO 99/34833; WO 99/14259, and U.S. Pat. No. 6,348,558.
In certain embodiments, macromolecular polymer (e.g., PEG) is attached to an agent described herein through an intermediate linker. In certain embodiments, the linker is made up of from 1 to 20 amino acids linked by peptide bonds, wherein the amino acids are selected from the 20 common, naturally-occurring amino acids. Some of these amino acids may be glycosylated, as is well understood by those in the art. In other embodiments, the 1 to 20 amino acids are selected from glycine, alanine, proline, asparagine, glutamine, and lysine. In other embodiments, a linker is made up of a majority of amino acids that are sterically unhindered, such as glycine and alanine. Non-peptide linkers are also possible. For example, alkyl linkers such as —NH(CH2)nC(O)—, wherein n=2-20 can be used. These alkyl linkers may further be substituted by any non-sterically hindering group such as lower alkyl (e.g., C1-C6) lower acyl, halogen (e.g., Cl, Br), CN, NH2, phenyl, etc. U.S. Pat. No. 5,446,090 describes a bifunctional PEG linker and its use in forming conjugates having a peptide at each of the PEG linker termini.
The peptides can also be modified, e.g., to further facilitate cellular uptake or increase in vivo stability, in some embodiments. For example, acylating or PEGylating a peptidomimetic macrocycle facilitates cellular uptake, increases bioavailability, increases blood circulation, alters pharmacokinetics, decreases immunogenicity and/or decreases the needed frequency of administration.
One manner of making of the peptides described herein is using solid phase peptide synthesis (SPPS). The C-terminal amino acid is attached to a cross-linked polystyrene resin via an acid labile bond with a linker molecule. This resin is insoluble in the solvents used for synthesis, making it relatively simple and fast to wash away excess reagents and by-products. The N-terminus is protected with the Fmoc group, which is stable in acid, but removable by base. Any side chain functional groups are protected with base stable, acid labile groups.
Peptide bonds can be replaced, e.g., to increase physiological stability of the peptide, by: a retro-inverso bonds (C(O)—NH); a reduced amide bond (NH—CH2); a thiomethylene bond (S—CH2 or CH2—S); an oxomethylene bond (O—CH2 or CH2—O); an ethylene bond (CH2—CH2); a thioamide bond (C(S)—NH); a trans-olefin bond (CH═CH); a fluoro substituted trans-olefin bond (CF═CH); a ketomethylene bond (C(O)—CHR) or CHR—C(O) wherein R is H or CH3; and a fluoro-ketomethylene bond (C(O)—CFR or CFR—C(O) wherein R is H or F or CH3.
The polypeptides can be further modified by: acetylation, amidation, biotinylation, cinnamoylation, farnesylation, fluoresceination, formylation, myristoylation, palmitoylation, phosphorylation (Ser, Tyr or Thr), stearoylation, succinylation and sulfurylation. As indicated above, peptides can be conjugated to, for example, polyethylene glycol (PEG); alkyl groups (e.g., C1-C20 straight or branched alkyl groups); fatty acid radicals; and combinations thereof.
In some instances, peptides can include a detectable label (a moiety that has at least one element, isotope, or functional group incorporated into the moiety which enables detection of the peptide to which the label is attached). Labels can be directly attached (i.e., via a bond) or can be attached by a linker (e.g., such as, for example, a cyclic or acyclic, branched or unbranched, substituted or unsubstituted alkylene; cyclic or acyclic, branched or unbranched, substituted or unsubstituted alkenylene; cyclic or acyclic, branched or unbranched, substituted or unsubstituted alkynylene; cyclic or acyclic, branched or unbranched, substituted or unsubstituted heteroalkylene; cyclic or acyclic, branched or unbranched, substituted or unsubstituted heteroalkenylene; cyclic or acyclic, branched or unbranched, substituted or unsubstituted heteroalkynylene; substituted or unsubstituted arylene; substituted or unsubstituted heteroarylene; or substituted or unsubstituted acylene, or any combination thereof, which can make up a linker). Labels can be attached to a peptide at any position that does not interfere with the biological activity or characteristic of the inventive polypeptide that is being detected.
Labels can include: labels that contain isotopic moieties, which may be radioactive or heavy isotopes, including, but not limited to, 2H, 3H, 13C, 14C, 15N, 31P, 32P 35S, 67Ga, 99mTc (Tc-99m), 111In, 123I, 125I, 169Yb and 186Re; labels that include immune or immunoreactive moieties, which may be antibodies or antigens, which may be bound to enzymes {e.g., such as horseradish peroxidase); labels that are colored, luminescent, phosphorescent, or include fluorescent moieties (e.g., such as the fluorescent label FITC); labels that have one or more photoaffinity moieties; labels that have ligand moieties with one or more known binding partners (such as biotin-streptavidin, FK506-FKBP, etc.).
In some instances, labels can include one or more photoaffinity moieties for the direct elucidation of intermolecular interactions in biological systems. A variety of known photophores can be employed, most relying on photoconversion of diazo compounds, azides, or diazirines to nitrenes or carbenes (see, e.g., Bayley, H., Photogenerated Reagents in Biochemistry and Molecular Biology (1983), Elsevier, Amsterdam, the entire contents of which are incorporated herein by reference). In certain embodiments of the invention, the photoaffinity labels employed are o-, m- and p-azidobenzoyls, substituted with one or more halogen moieties, including, but not limited to 4-azido-2,3,5,6-tetrafluorobenzoic acid.
Labels can also be or can serve as imaging agents. Exemplary imaging agents include, but are not limited to, those used in positron emissions tomography (PET), computer assisted tomography (CAT), single photon emission computerized tomography, x-ray, fluoroscopy, and magnetic resonance imaging (MRI); anti-emetics; and contrast agents. Exemplary diagnostic agents include but are not limited to, fluorescent moieties, luminescent moieties, magnetic moieties; gadolinium chelates (e.g., gadolinium chelates with DTPA, DTPA-BMA, DOTA and HP-D03A), iron chelates, magnesium chelates, manganese chelates, copper chelates, chromium chelates, iodine-based materials useful for CAT and x-ray imaging, and radionuclides. Suitable radionuclides include, but are not limited to, 123I, 125I, 130I, 131I, 133I, 135I, 47Sc, 72As, 72Se, 90Y, 88Y, 97Ru, 100Pd, 101mRh, 119Sb, 128Ba, 197Hg, 211At, 212Bi, 212Pb, 109Pd, 111In, 67Ga, 68Ga, 67Cu, 75Br, 77Br, 99mTc, 14C, 13N, 15O, 32P, 33P, and 18F.
Fluorescent and luminescent moieties include, but are not limited to, a variety of different organic or inorganic small molecules commonly referred to as “dyes,” “labels,” or “indicators.” Examples include, but are not limited to, fluorescein, rhodamine, acridine dyes, Alexa dyes, cyanine dyes, etc. Fluorescent and luminescent moieties may include a variety of naturally-occurring proteins and derivatives thereof, e.g., genetically engineered variants. For example, fluorescent proteins include green fluorescent protein (GFP), enhanced GFP, red, blue, yellow, cyan, and sapphire fluorescent proteins, reef coral fluorescent protein, etc. Luminescent proteins include luciferase, aequorin and derivatives thereof. Numerous fluorescent and luminescent dyes and proteins are known in the art (see, e.g., U.S. Patent Publication 2004/0067503; Valeur, B., “Molecular Fluorescence: Principles and Applications,” John Wiley and Sons, 2002; and Handbook of Fluorescent Probes and Research Products, Molecular Probes, 9th edition, 2002).
Pharmaceutical Compositions
One or more of the peptides disclosed herein (e.g., one or more of SEQ ID NOs: 1-12) can be formulated for use as or in pharmaceutical compositions. Such compositions can be formulated or adapted for administration to a subject via any route, e.g., any route approved by the Food and Drug Administration (FDA). Exemplary methods are described in the FDA's CDER Data Standards Manual, version number 004 (which is available at fda.give/cder/dsm/DRG/drg00301.htm). For example, compositions can be formulated or adapted for administration by inhalation (e.g., oral and/or nasal inhalation (e.g., via nebulizer or spray)), injection (e.g., intravenously, intra-arterial, subdermally, intraperitoneally, intramuscularly, and/or subcutaneously); and/or for oral administration, transmucosal administration, and/or topical administration (including topical (e.g., nasal) sprays and/or solutions).
In some instances, pharmaceutical compositions can include an effective amount of one or more peptides. The terms “effective amount” and “effective to treat,” as used herein, refer to an amount or a concentration of one or more compounds or a pharmaceutical composition described herein utilized for a period of time (including acute or chronic administration and periodic or continuous administration) that is effective within the context of its administration for causing an intended effect or physiological outcome (e.g., treatment of cancer).
Pharmaceutical compositions of this invention can include one or more peptides and any pharmaceutically acceptable carrier and/or vehicle. In some instances, pharmaceuticals can further include one or more additional therapeutic agents in amounts effective for achieving a modulation of disease or disease symptoms.
The term “pharmaceutically acceptable carrier or adjuvant” refers to a carrier or adjuvant that may be administered to a patient, together with a compound of this invention, and which does not destroy the pharmacological activity thereof and is nontoxic when administered in doses sufficient to deliver a therapeutic amount of the compound.
The pharmaceutical compositions may contain any conventional non-toxic pharmaceutically-acceptable carriers, adjuvants or vehicles. In some cases, the pH of the formulation may be adjusted with pharmaceutically acceptable acids, bases or buffers to enhance the stability of the formulated compound or its delivery form. The term parenteral as used herein includes subcutaneous, intra-cutaneous, intra-venous, intra-muscular, intra-articular, intra-arterial, intra-synovial, intra-sternal, intra-thecal, intra-lesional and intra-cranial injection or infusion techniques.
A sterile injectable preparation may also be a sterile injectable solution or suspension in a non-toxic parenterally acceptable diluent or solvent, for example, as a solution in 1,3-butanediol. Among the acceptable vehicles and solvents that may be employed are mannitol, water, Ringer's solution and isotonic sodium chloride solution. In addition, sterile, fixed oils are conventionally employed as a solvent or suspending medium. For this purpose, any bland fixed oil may be employed including synthetic mono- or diglycerides. Fatty acids, such as oleic acid and its glyceride derivatives are useful in the preparation of injectables, as are natural pharmaceutically-acceptable oils, such as olive oil or castor oil, especially in their polyoxyethylated versions. These oil solutions or suspensions may also contain a long-chain alcohol diluent or dispersant, or carboxymethyl cellulose or similar dispersing agents which are commonly used in the formulation of pharmaceutically acceptable dosage forms such as emulsions and or suspensions.
In some instances, one or more peptides disclosed herein can be conjugated, for example, to a carrier protein. Such conjugated compositions can be monovalent or multivalent. For example, conjugated compositions can include one peptide disclosed herein conjugated to a carrier protein. Alternatively, conjugated compositions can include two or more peptides disclosed herein conjugated to a carrier.
As used herein, when two entities are “conjugated” to one another they are linked by a direct or indirect covalent or non-covalent interaction. In certain embodiments, the association is covalent. In other embodiments, the association is non-covalent. Non-covalent interactions include hydrogen bonding, van der Waals interactions, hydrophobic interactions, magnetic interactions, electrostatic interactions, etc. An indirect covalent interaction is when two entities are covalently connected, optionally through a linker group.
Carrier proteins can include any protein that increases or enhances immunogenicity in a subject. Exemplary carrier proteins are described in the art (see, e.g., Fattom et al., Infect. Immun., 58:2309-2312, 1990; Devi et al., Proc. Natl. Acad. Sci. USA 88:7175-7179, 1991; Li et al., Infect. Immun. 57:3823-3827, 1989; Szu et al., Infect. Immun. 59:4555-4561, 1991; Szu et al., J. Exp. Med. 166:1510-1524, 1987; and Szu et al., Infect. Immun. 62:4440-4444, 1994). Polymeric carriers can be a natural or a synthetic material containing one or more primary and/or secondary amino groups, azido groups, or carboxyl groups. Carriers can be water soluble.
Methods of Treatment
The disclosure includes methods of using the peptides herein for the prophylaxis and/or treatment of cancer. The terms “treat” or “treating,” as used herein, refers to partially or completely alleviating, inhibiting, ameliorating, and/or relieving the disease or condition from which the subject is suffering.
In general, methods include selecting a subject and administering to the subject an effective amount (a therapeutically effective amount) of one or more of the peptides herein, e.g., in or as a pharmaceutical composition, and optionally repeating administration as required for the prophylaxis or treatment of a cancer.
Specific dosage and treatment regimens for any particular patient will depend upon a variety of factors, including the activity of the specific compound employed, the age, body weight, general health status, sex, diet, time of administration, rate of excretion, drug combination, the severity and course of the disease, condition or symptoms, the patient's disposition to the disease, condition or symptoms, and the judgment of the treating physician.
Described below are studies that identified a 36-residue human proteome-derived peptide binders of the ENAH EVH1 domain. The studies identified three distinct ways in which local and distal sequence elements surrounding the FP4 motif modulate ENAH binding affinity and specificity. Also identified a network of dispersed residues in the ENAH EVH1 domain that allows it to adopt a conformation inaccessible to EVL and VASP, allowing ENAH to selectively bind to retinal protein PCARE. In the studies described below, this interaction were validated in cells. The studies below also describe a nanomolar EVH1 binder that is 400-600-fold selective for ENAH over VASP and EVL.
MassTitr is a recently developed SORT-SEQ method, similar to Tite-seq, that is based on fluorescence-activated cell sorting (FACS) of a library of peptide-displaying bacterial and subsequent deconvolution of signals by deep sequencing. Specifically, as shown in
We used MassTitr to identify ENAH EVH1 binders from the human proteome by screening a library of 416,611 36-mer peptides with 7-residue overlaps (the T7-pep library). The T7-pep library spans the entire protein-coding space of the human genome (
We validated 16 peptide hits for binding to monomeric ENAH EVH1 domain, using biolayer interferometry (BLI) to determine dissociation constants that ranged from 0.19 μM to 63 μM (Table 2). With the exception of SHROOM3 peptide, binders classified as high-affinity by MassTitr bound to ENAH EVH1 domain more tightly than peptides classified as low-affinity. Many newly identified peptides bound with affinities similar to or tighter than a well-studied control peptide from Listeria monocytogenes protein ActA, which bound with KD=5.2 μM in our BLI assay (Table 2). Prior to this work, a single FP4-motif containing sequence from ActA was the tightest known endogenously derived binder of Ena/VASP EVH1 domains.
Although the majority of MassTitr hits contained FP4 motifs (
The remaining 35 noncanonical sequences contained CXC motifs. These sequences bound robustly to the ENAH tetramer in bacterial display, and mutating either of the cysteines in a CXC-containing peptide from OLIG3 abrogated binding to ENAH (
MassTitr detects ENAH EVH1 domain interactions with short peptides displayed on the cell surface. To highlight putative biologically relevant interaction partners of ENAH, we applied a bioinformatic analysis to identify those motifs that are likely to be accessible and co-localized with ENAH. We filtered our high-confidence hits by disorder propensity (IUPRED>0.4) and cytoplasmic subcellular localization. This resulted in 33 peptides, of which 15 are derived from interaction partners known to interact or co-localize with Ena/VASP proteins. Filtered hits were highly enriched in GO biological process terms including actin filament organization (FDR<10−6), and positive regulation of cytoskeleton (FDR<0.05), which align with documented cellular functions of ENAH. Notably, we also identified proteins localized to the Golgi body and cilia, where Ena/VASP function is less defined (
Next, we turned to identifying FP4 SLiM-flanking elements that enhance binding to the ENAH EVH1 domain. A sequence logo made of the high-confidence MassTitr hits shows enrichment of prolines C-terminal to the FP4 motif, and a binomial test confirms that peptides containing FP4 motifs followed by three consecutive prolines are enriched our hit list (p<10−11;
To understand the structural basis of high-affinity ABI1 binding by ENAH EVH1 domain, we solved a crystal structure of the complex at 1.88 Å resolution. Only the FP8 region was fully resolved in the structure under the crystallization conditions, which included high salt and low pH (
Another enriched feature of MassTitr-identified binders, relative to the pre-screened input library, is the presence of multiple FP4 motifs (binomial test, p<10−22). Multiple motifs are also enriched in MassTitr high-affinity hits relative to all hits (p<0.02), supporting the idea that multiple FP4 motifs enhance affinity. Peptide binders with multiple motifs have preferred spacings of approximately 5 or 15 residues between FP4 motifs (
It was previously shown that zyxin, which contains four clustered FP4 motifs, binds to the VASP EVH1 domain by contacting both the canonical FP4 site and a secondary, noncanonical binding site on the opposite side of the EVH1 domain. To examine whether multi-FP4 peptides may achieve higher affinity by engaging an analogous secondary binding site on ENAH, we designed experiments to disrupt this putative site. A crystal structure of ENAH bound to an ActA peptide (PDB 5NC7) contains a second ActA chain bound to the region corresponding to the noncanonical binding site. This was presumed to be a packing artifact, but we speculated that a similar interaction might stabilize the binding of dual-motif ligands. In the ActA peptide-bound structure, the ENAH Arg47 side chain forms a hydrogen bond with a carbonyl on the PPII helix backbone of ActA. In support of a role for this residue in noncanonical site binding, the analogous VASP Arg48 exhibits significant NMR HSQC chemical shifts upon titration with zyxin peptide. Thus, we tested binding of multi-FP4 motif-containing peptides to ENAH EVH1 domain with Arg47 mutated to Ala (ENAH R47A). While the affinities of single-FP4 motif peptides from ActA and PCARE were minimally affected by this mutation, peptides containing two FP4 motifs from zyxin, LPP, and NHSL1 experienced a 7-15-fold reduction in affinity to ENAH R47A relative to wild type, suggesting that these multi-FP4 peptides gain affinity by contacting the noncanonical site (Table 4,
We probed the extent to which binding affinity depended on the presence of multiple FP4 motifs. A 36-mer sequences from LPP and NHSL1 contain two FP4 motifs, and in both cases, truncating the peptide to leave only one motif reduced affinity 3-6-fold. Interestingly, the identity of the sequence between the two motifs is also important, as replacing this with a Gly-Ser linker in LPP led to a ˜4-fold reduction in affinity (Table 1,
Using isothermal titration calorimetry (ITC), we confirmed that dual-motif peptides from LPP and NHSL1 and the single FP4 motif peptide from ActA fit well to a 1:1 binding model (
Finally, we examined the motif-spacing requirements of bidentate binding. To estimate the minimum number of residues required to link a peptide bound in the canonical site to one in the noncanonical site, we generated a structure of the ENAH EVH1 domain bound to two FP4 peptides, using the binding poses in our ABI1-bound structure and structure 5NC7. We then used Rosetta to build a peptide chain to connect the two peptides. There are two different orientations for a peptide that links the FP4 motifs, depending on which site is bound by the N-terminal motif, as shown in
FPPPPPPPP
FPPPPSSSS
PPPPMEVLMDKSFASLES
PAPLEEEIFPSPPPPPEE
PTEDELEIIRETASSLDS
The highest affinity peptide that we discovered was from PCARE, a retinal protein (KD=0.19 μM for 36-residue peptide PCARE813-848; Table 2). Successive truncations of this peptide identified the minimal region for high affinity binding (PCARE826-848, which we call PCARE B, KD=0.32 μM,
PCARE residues Phe843 and Leu846 make hydrophobic interactions with Ala83 and Pro65 on ENAH. The side chain of Asp840 on PCARE docks into a polar pocket on ENAH made up of the backbone amines of ENAH Leu68, Lys69, and Arg81. Notably, the backbone amine and side chain of Ser842 on PCARE forms hydrogen bonds with the side chain of Asp840, correctly positioning Asp840 to hydrogen bond with a water molecule that is coordinated by the backbone amine of ENAH Arg81. Most intriguing is the interaction between Val837 on PCARE and the hydrophobic groove in ENAH that typically binds to FP4 motifs. Rather than using a large aromatic such as Phe to dock into this pocket, the alpha helical structure of PCARE positions the smaller Val837 in this site in an orientation entirely different from the way a canonical Phe binds in other complexes (
Next, we examined whether residues flanking FP4 SLiMs impact paralog specificity. Surprisingly, we found that PCARE B had a 70-140-fold preference for binding to ENAH over the VASP or EVL EVH1 domains (
When ENAH EVH1 domain binds to PCARE828-848, a loop in ENAH (residues 80-86) shifts 3 Å relative to the ENAH bound to the ligand FPPPP (1EVH), leading to compression of ENAH's hydrophobic core. EVL and VASP have bulkier residues in this region, and modeling on the ENAH-PCARE828-848 structure predicts severe steric clashes if EVL or VASP were to adopt the PCARE828-848-bound conformation observed for ENAH (
To identify the network of residues responsible for ENAH's binding specificity, we used the structure-based modeling method dTERMen (design with TERM energies). dTERMen is a protocol for scoring the compatibility of a sequence with a structure based on the frequencies with which combinations of residues are found in compact tertiary motifs in known structures. We first scored the native EVL sequence on the PCARE828-848-bound ENAH template with dTERMen, which scored considerably worse than the native ENAH sequence. Guided by the dTERMen score, we introduced increasing numbers of residues from ENAH into the sequence of EVL. Six replacements were sufficient to recapitulate dTERMen energies similar to that for ENAH in the PCARE828-848-bound conformation. We made a mutated version of EVL with the 6 corresponding residues from ENAH, also including the I26A mutation in EVL as mentioned above. The replaced residues are distributed across the EVH1 domain, and several are distant from the PCARE828-848 binding site (
To test whether a PCARE-derived peptide could selectively bind to ENAH in cells, we used Ena/VASP family-deficient cell line MVD7 (Bear 2000) and expressed individual paralogs ENAH, EVL, and VASP to evaluate interactions with PCARE. MVD7 are embryonic fibroblasts derived from Enah−/− Vasp−/− mice, with low Evl expression (Auerbach 2003, Damiano-Guercio 2020). As Ena/VASP family members are known to heterotetramerize (Riquelme 2015), we used an shRNA against Evl to provide a clean background (average knockdown of MVD7shEVL versus MVD7 42.8+/−1.8% by qPCR, 3 biological replicates). We cloned the PCARE peptide sequence (PCARE822-848) into a lentiviral expression vector, tagging the N-terminus with mRuby2 and a disordered linker to enable visualization (mRuby2-PCARE). mRuby2-PCARE was co-expressed with GFP, GFP-tagged ENAH, an EVH1 deletion mutant (ΔEVH1-ENAH), EVL, or VASP. ENAH, EVL, and VASP all robustly localized to focal adhesions, while GFP and ΔEVH1-ENAH were largely cytoplasmic. mRuby2-PCARE exhibited diffuse cytoplasmic localization under all conditions except when co-expressed with ENAH, in which case mRuby2-PCARE was moderately recruited to focal adhesions (
To determine if PCARE is capable of disrupting ENAH localization and to further explore the PCARE-ENAH interaction in cells, we tagged PCARE822-848 with a mitochondrial localization sequence, generating Mito-mRuby2-PCARE. In MVD7shEVL cells with reconstituted Ena/VASP paralogs, exogenously expressed ENAH was strongly enriched at both focal adhesions and at mitochondria, while EVL and VASP remained entirely localized to focal adhesions (
Finally, we sought to determine whether expression of cytosolic or mitochondria-targeted PCARE822-848 peptide could disrupt ENAH-mediated cell adhesion. We examined early focal adhesion maturation in the breast cancer cell line MCF7 (
Based on our results demonstrating a PCARE-derived peptide could selectively recruit and inhibit ENAH-dependent functions in cells, we sought to engineer higher affinity peptides to be used either as inhibitors against ENAH, an attractive therapeutic target for cancer metastasis, or as tools to study Ena/VASP biology. We integrated what we learned about peptide features that promote binding of native proteins to ENAH and took a rational design approach to making even tighter-binding designed peptides. With the goal of combining peptide elements to target multiple sites along the ENAH EVH1 domain (
Materials and Methods for the Examples
FACS Sample Preparation and Analysis
The general protocol for sample preparation for FACS analysis and Mass-Titr sorting was as follows: 5 mL cell cultures of eCPX plasmid expressing either library or control peptide was grown overnight at 37° C. in LB+25 μg/ml chloramphenicol+0.2% w/v glucose. The next day, the O.D. 600 of the cultures were measured. Enough cells for 100 L of an O.D. 600 of 6.0 were pelleted and inoculated into fresh TB+25 μg/mL chloramphenicol and grown at 37° C. Upon reaching an O.D. 600 of 0.5-0.6, cells were induced with 0.04% w/v arabinose and induced for 1.5 hours at 37° C. The O.D. 600 was then remeasured, and enough cells were pelleted for analysis (1×107 cells per FACS analysis sample, 7×107 cells for library sorting). Cells were resuspended to a concentration of 4×108 cells/mL, washed in PBS+0.1% BSA, and then incubated with anti-FLAG antibody conjugated to APC (DFLAG-APC; PerkinElmer) diluted 1:100 in PBS+0.1% BSA at a ratio of 30 μL labelled antibody:107 cells. Cells wrapped in foil were incubated at 4° C. for 15 minutes, washed with PBS+0.1% BSA, and pelleted. For each FACS analysis sample, 25 μL of 1×107 cells in PBS were mixed with 25 μL of a 2× concentration of Mena tetramer in PBS+1% BSA+4 mM DTT (final concentration of 2 mM DTT), and then incubated at 4° C. for 1 hour in foil. After incubation, 50 μL of the mixture was added per well to a 96-well Multi-Screen HTS GV sterile filtration plate (Millipore), buffer was removed by vacuum, and then cells were washed twice with 200 L of PBS+0.5% BSA. Each well containing 1×107 cells was then resuspended in 30 L of streptavidin-PE (SAV-PE; ThermoFisher Scientific) diluted 1:100 in PBS+0.1% BSA, and incubated for 15 minutes at 4° C., washed with 200 μL of PBS+0.1% BSA, and resuspended in 250 μL of PBS+0.1% BSA for subsequent FACS analysis or sorting.
Biolayer Interferometry (BLI)
All BLI experiments were performed with 2 replicates on an Octet Red96 instrument (ForteBio). Biotinylated, 6×-His-Sumo-peptide fusions purified in small scale were diluted into BLI buffer (PBS pH 7.4, 1% BSA, 0.1% Tween-20, 1 mM DTT) and immobilized onto streptavidin-coated tips (ForteBio) until loading reached a response level between 0.5-0.6 nm. The loaded tips were immersed in a solution of ENAH EVH1 domain at a relevant dilution series in BLI buffer at an orbital shake speed of 1000 rpm and run until the binding signal plateaued. ENAH-bound tips were subsequently placed into BLI buffer only for dissociation and run until the binding signal plateaued. KD values were obtained through steady state analysis. Briefly, the data were corrected for background by subtracting the signal obtained when doing the same experiment using a biotinylated, 6×-His Sumo instead of an immobilized ENAH-binding peptide. The association phases were then fit to a one-phase binding model in Prism and the equilibrium steady-state binding signal values from that fit were plotted against ENAH concentration and fit to a single-site specific binding model in Prism to obtain dissociation constants. Errors reported are the standard deviation of two replicates.
Isothermal Titration Calorimetry (ITC)
ITC experiments were performed with two replicates using a VP-ITC microcalorimeter (MicroCal LLC). To prepare samples for ITC, 2.5 mL of 100 μM ENAH EVH1 domain and 1 mL of 800 μM-1.2 mM of SUMO-peptide fusions were dialyzed against 2 L of ITC Buffer (20 mM Hepes pH 7.6, 150 mM NaCl, 1 mM TCEP) at 4° C. overnight. The concentrations of proteins were remeasured after dialysis on the day of experiment. SUMO-peptide was titrated into ENAH EVH1 domain at 25° C. Data analysis and curve fitting were performed with the Origin 7.0 software (OriginLab). Errors reported as the standard deviation of two replicates.
Crystallography
Crystals of ENAH fused to ABI1 were grown in hanging drops over a reservoir containing 0.1 M Sodium Acetate pH 4.5 and 2.90 M NaCl. 1 μL of ENAH-ABI1 (250 M in 20 mM Hepes, 150 mM NaCl, 1 mM DTT) was mixed with 1 μL of reservoir solution, and 3D crystals appeared in two weeks. Crystals of ENAH fused to PCARE were grown in hanging drops containing 0.1M Tris pH 8.0 and 3.30 M NaCl. 1.5 μL of ENAH-PCARE (769 μM in 20 mM Hepes, 150 mM NaCl, 1 mM DTT) were mixed with 0.5 μL of reservoir solution, and football shaped crystals appeared in two days. Both crystals were grown at room temperature. Diffraction data were collected at the Advanced Photon Source at Argonne National Laboratory, NE-CAT beamline 24-IDE. The ENAH-ABI1 and ENAH-PCARE integrated and scaled to 1.88 Å and 1.65 Å with HKL2000 and the structures were solved with molecular replacement using either ENAH EVH1 structures 5NC7 or 6RD2 as search models respectively. Structures were refined with iterative rounds of model rebuilding with PHENIX and COOT.
Cell Culture
MCF7 and HEK293T cells were cultured in high-glucose Dulbecco's modified Eagle's medium (DMEM) base media with sodium pyruvate (Corning) supplemented with 10% fetal bovine serum (FBS; Millipore), 2 mM L-glutamine (Corning), and 100 U/mL penicillin with 100 μg/mL streptomycin (Corning). MVD7 cells were cultured in high-glucose DMEM supplemented with 15% FBS, 2 mM L-glutamine, 100 U/mL mouse interferon-T (Millipore). MCF7 and HEK293T cells were maintained in a 37° C. humidified incubator under 5% CO2. MVD7 cells were maintained in a 32° C. humidified incubator under 5% CO2. For lentiviral production, second generation lentiviral particles were generated by PEI transfection of 293T cells with transfer plasmid, pMD2.G, and psPAX2 (Addgene #12259, #12260, gifts from Didier Trono). HEK293T media containing lentiviral particles was collected, filtered, and added directly to cultures with polybrene (Gibco). Puromycin (2 μg/mL final concentration; Thermo Fisher) and blasticidin (4 g/mL final; Gibco) were used to select for cells stably expressing shRNA sequences or Ena/VASP constructs, respectively, after lentiviral transduction. For immunofluorescence experiments, MCF7 cells were cultured on fibronectin-coated coverslips (10 μg/mL; Corning). For live-cell imaging (MVD7), cells were cultured in fibronectin-coated glass-bottom dishes (Mattek).
Reverse Transcription Quantitative PCR (RT-qPCR)
To assess Evl knockdown in MVD7, total cellular RNA was isolated using the Isolate II RNA kit (Bioline) according to manufacturer's instructions. cDNA was synthesized from 1000 ng of input RNA using qScript cDNA Synthesis kit (Quantabio). RT-qPCR reactions were run in duplicate on an ABI 7500 Fast Real-Time PCR System (Applied Biosystems) with PowerTrack SYBR Green Master Mix (Thermo Fisher). Primer pairs were confirmed to have 85-110% efficiency based on the slope of the standard curve from a cDNA dilution series. CTS were normalized to the CT GAPDH housekeeping genes. Percent knockdown was determined using the comparative CT method. Mus musculus Gapdh Fwd AGGTCGGTGTGAACGGATTTG, Rev GGGGTCGTTGATGGCAACA. Evl Fwd TGAGAGCCAAACGGAAGACC, Rev TTCTGGACAGCAACGAGGAC.
Western Blotting
Cells were lysed in buffer containing 140 mM NaCl, 10 mM Tris pH 8.0, 1 mM EDTA, 0.5 mM EGTA, 1% Triton X-100, 0.1% sodium deoxycholate, and 0.1% SDS with protease and phosphatase inhibitors (Boston Bio Products). Samples were resolved by SDS-PAGE and transferred onto nitrocellulose membranes. Membranes were blocked in Odyssey Blocking Buffer (LI-COR) for 1 hour and incubated at 4° C. overnight with primary antibodies. Primary antibodies were used as follows: mouse Actin 1:2500 (ProteinTech Group, 66009-1-Ig), mouse GAPDH 1:1000 (Cell Signaling Technology, 5174S), rabbit ENAH 1:250 (Sigma, HPA028696), rabbit EVL 1:1000 (Sigma, HPA018849), rabbit VASP 1:1000 (Cell Signaling Technology, 3132S). Membranes were incubated with secondary antibodies conjugated to either Alexa Fluor 680 or 790 (Thermo Fisher) for 1 hour. Immunoblots were scanned using Odyssey CLx imager (LI-COR).
Immunofluorescence
MCF7 cells were fixed and immunolabeled 8 hours after plating onto fibronectin (10 g/mL; Corning) coated coverslips to assay focal adhesions. Cells were fixed with 4% paraformaldehyde (PFA; Electron Microscopy Services) with 0.075 mg/mL saponin (Alfa Aesar, Sigma) diluted in phosphate-buffered saline (PBS) at 37° C. for 10 minutes. PFA was quenched with 100 mM glycine in PBS at room temperature for 10 minutes. Cells were then blocked in 1% BSA and 1% FBS in PBS either overnight at 4° C. or for 1 hour at room temperature. The following immunofluorescence reagents and antibodies were used: mouse anti-Paxillin (1:200, clone: 349; BD Biosciences, 612405), rabbit anti-ENAH (1:100; HPA028448, Sigma), goat anti-mouse Alexa Fluor 488 and Alexa Fluor 647 (1:1000, Thermo Fisher), donkey anti-rabbit Alexa Fluor 405 (Abcam), and Alexa Fluor 488- and Alexa Fluor 647-Phalloidin (1:40, Thermo Fisher). Primary antibodies were diluted in block solution and incubated for 1.5 hours at room temperature. After washing, coverslips were incubated for 1 hour in secondary antibody solution with fluorescently-labelled phalloidin. Coverslips were mounted using ProLong Gold Antifade (Invitrogen) and allowed to cure for at least 24 hours before imaging.
Microscopy and Image Analysis
Cells were imaged on a Ti-E inverted microscope (Nikon), with a 100×Apo TIRF 1.49 NA objective (Nikon) and an ORCA-Flash 4.0 V2 CMOS camera (Hamamatsu). For focal adhesion assessment, cells were imaged with total internal reflection fluorescence (TIRF) microscopy. To increase the depth of imaging for examination of mitochondrial localization, standard widefield fluorescence microscopy was used. Focal adhesion quantification was performed as previously described (Puleo 2019). Briefly, a binary mask was generated for paxillin signal and actin signal, denoting focal adhesions and cell area, respectively. To facilitate semi-automated segmentation of focal adhesions, we generated a sharp, high contrast image of the paxillin and actin channels by the following processing steps: deconvolution using 5 iterations of the Richardson-Lucy algorithm, shading correction using rolling ball, and unsharp masked (NIS Elements). Focal adhesion area was quantified by measuring the paxillin area of each cell within the whole cell area, or by examining individual focal adhesions. Each experimental condition was performed in triplicate and plotted together. Images presented in figures have been lightly processed in NIS Elements, including by applying 2 iterations of the Richardson-Lucy deconvolution algorithm and rolling ball shading correction to reduce background in live cell images.
Protein Constructs Purification:
Human ENAH EVH1 domain, followed by a 6×-Gly linker and ENAH mouse coiled coil (for tetramerization), were cloned into a pDW363 biotinylation vector that includes a C-terminal biotin acceptor peptide (BAP) tag and a 6×-His tag. This ENAH tetramer construct was expressed in Rosetta2(DE3) (Novagen) cells in Terrific Broth (TB) with 100 μg/mL ampicillin, 25 μg/mL chloramphenicol, and 0.050 mM D-(+)-biotin (for in vivo biotinylation). Cells were grown at 37° C. with shaking to an optical density at 600 nm (O.D. 600) of 0.5-0.7 and then induced with 1 mM IPTG and grown at 37° C. for 5 hours. 1 L of cells were then spun down and resuspended in 25 mL of binding buffer (20 mM Tris pH 8.0, 500 mM NaCl, 5 mM imidazole), and frozen at −80° C. overnight. The next day, pellets were thawed and supplemented with 20 mM phenylmethylsulfonyl fluoride (PMSF) protease inhibitor. Cells were sonicated ten times for 30 s followed by 30 s of rest on ice and centrifuged. The clarified lysate was filtered through a 0.2 μm filter and applied to 2 mL of Ni-nitrilotriacetic (Ni-NTA) acid agarose resin (GoldBio) equilibrated in wash buffer (20 mM Tris pH 8.0, 500 mM NaCl, 20 mM imidazole). The resin was then washed 3 times with 8 mL wash buffer and eluted with 10 mL of elution buffer (20 mM Tris pH 8.0, 500 mM NaCl, 300 mM imidazole). The elution was run through a S75 26/60 size exclusion column equilibrated in gel filtration buffer (20 mM Tris pH 8.0, 150 mM NaCl, 1 mM DTT, 10% glycerol). Purity was verified by SDS-PAGE, and the fractions were pooled, concentrated, and flash frozen at −80° C.
Monomeric ENAH, EVL, or VASP EVH1 domains for use in BLI and ITC experiments, or ENAH EVH1-peptide fusions for crystallography was cloned into a pMCSG7 vector (gift from Frank Gertler, MIT), which places a 6×-His and TEV cleavage tag N-terminal to the EVH1 domain. These constructs were transformed into Rosetta2(DE3) cells and grown in 2×YT media supplemented with 100 μg/mL ampicillin. Cells were grown while shaking at 37° C. to an O.D. 600 of 0.5-0.7, and then cooled on ice for at least 20 minutes. They were then induced with 0.5 mM IPTG and grown while shaking at 18° C. overnight. Induced cultures were resuspended in 25 mL of wash buffer (20 mM Hepes pH 7.6, 500 mM NaCl, 20 mM imidazole) and frozen at −80° C. overnight. The next day, cultures were sonicated, spun down, and applied to Ni-NTA acid agarose resin equilibrated with wash buffer, and washed as described above. Samples were eluted in 10 mL elution buffer (20 mM Hepes pH 7.6, 500 mM NaCl, 30 mM imidazole). TEV protease was added to the elution at a ratio of 1 mg TEV:50 mg tagged protein along with 1 mM DTT. This mixture was dialyzed against TEV cleavage buffer (50 mM Hepes pH 8.0, 300 mM NaCl, 5 mM DTT, 1 mM EDTA) at 4° C. overnight and then applied over Ni-NTA acid agarose resin equilibrated with wash buffer. The column was washed with 2×8 mL of wash buffer, and the resulting flow-through and washes were pooled, concentrated, and applied to an S75 26/60 column equilibrated in gel filtration buffer (20 mM Hepes pH 7.6, 150 mM NaCl, 1 mM DTT, 1% glycerol). Purity was verified by SDS-PAGE and combined fractions were concentrated and flash frozen at −80° C.
SUMO-peptide fusions were cloned into a pDW363 vector that appends a BAP and 6×-His tag to the N-terminus of the protein and transformed into Rosetta2(DE3) cells. For ITC experiments, these cells were expressed in TB supplemented with 100 μg/mL ampicillin, grown to an O.D. 600 of 0.5-0.7, and induced with 1 mM IPTG. Induced cultures were purified as described above for the pMCGS7 constructs, with the exception of the TEV cleavage step. Instead, after elution with elution buffer, the sample was directly applied to the S75 26/60 column equilibrated in gel filtration buffer. Fractions were pooled, concentrated, and flash frozen at −80° C.
Small Scale Protein Purification for Biolayer Interferometry
Rosetta2(DE3) cells (Novagen) were transformed with pDW363 vector encoding 6×-His-SUMO-peptide fusions and grown in 20 mL of LB+100 μg/mL Ampicillin+0.050 mM D-(+)-Biotin dissolved in DMSO for in vivo biotinylation. Cells were grown to an OD of 0.5-0.7 with 1 mM IPTG and induced for 4-6 hours at 37° C. Pellets were spun down and frozen and −80° C. for at least 2 hours. These pellets were then thawed and resuspended in B-PER reagent (ThermoFisher) at 4 mL/gram of pellet with 0.2 mM PMSF. This suspension was shaken at 25° C. for 10-15 minutes, and then spun down at 15,000 g for 10 minutes. The supernatant was applied to 250 μL of Ni-NTA acid agarose resin equilibrated in 20 mM Tris pH 8.0, 500 mM NaCl, 20 mM Imidazole (Buffer A), and then washed 3 times with 1 mL of this buffer. Peptides were eluted in 1.8 mL of 20 mM Tris pH 8.0, 500 mM NaCl, 300 mM imidazole to use in BLI assays.
Bacterial Cell Surface Display Plasmids and T7-Pep Library Cloning
Control peptides for display were expressed at the C-terminus of eCPX in a vector designed by the Daughtery group. This construct was modified to include a FLAG tag at the N-terminus of the peptide and a C-myc tag at the C-terminus. The T7-pep library plasmids (gift from Elledge lab, Harvard), were transformed into Pir1 cells (Thermofisher), grown, and miniprepped (Qiagen) to isolate the library plasmid. Plasmids were then cut with EcoRI (NEB) and XhoI (NEB), and the inserts were gel purified, combined and concentrated with a Zymo Clean and Concentrate column, and eluted with 50 μL of sterile MilliQ Water. To clone the T7-pep library into the eCPX vector, we first grew up 200 mL of an empty eCPX vector in DH5a cells at 37° C. overnight. This culture was miniprepped (Qiagen) and then digested with EcoI and XhoI at a ratio of 10 units of enzyme:1 μg of vector at 37° C. for 2 hours. The resulting digest was PCR purified and eluted with 40 μL water. This cut vector was then dephosphorylated with Antarctic phosphatase (NEB) at a ratio of 1 μL/1 μg of DNA at 37° C. for 2 hours, followed by 10 min at 65° C. for enzyme inactivation. T7-pep library insert was ligated into the cut eCPX vector using T4 ligase (NEB) at 14° C. overnight. Ligase was subsequently deactivated for 10 minutes at 70° C., and the ligation reaction was concentrated with Zymo Clean and Concentrate columns (Zymo Research). Each column was eluted with 12.5 μL elution buffer (from kit). The resulting elutions were desalted on a 0.025 μm filter (Millipore) for 15-20 minutes and pooled on ice. Electrocompetent MC1061 cells and 10-20 μL DNA were then mixed and transferred to a cold 2 mm cuvette (BioRad). Each cuvette was pulsed at 2.5 kV, 50 μF, 100 ohms on an electroporator (BioRad), immediately rinsed out with 3×1 mL of warm SOC and transferred to a culture tube containing 7 mL warm SOC. Cells were incubated at 37° C. for one hour, and then combined. Serial dilutions of the library were plated on LB/chloramphenicol plates to assess transformation efficiency, and the leftover cells were added to 500 mL of LB+25 μg/mL chloramphenicol+0.2% w/v sterile-filtered glucose. The library was grown at 37° C. until it reached an O.D. 600 of 2.0 and then frozen as glycerol stocks to use for FACS analysis and sorting.
All other peptides used in the experiment were cloned into the C-terminus of the following sequence:
Peptide sequences inserted after the C-terminus are listed in Tables 1, S2, and S3
Bolded sequences IPLR and RIAR constitute part of the EcoRI and XhoI restriction enzyme sites, respectively.
Peptide sequences are as follows:
Pre-Enrichment of T7-Pep Library
The T7-pep library was prepared as described above. 7×107 cells expressing the T7-pep library were incubated with a final concentration of 20 μM ENAH tetramer. Cells were sorted on a BDFACS Aria machine. Prior to sorting, a positive (ActA) control and a negative (empty) control were analyzed. A gate to collect cells expressing peptide binders was set that included 0.3% of the negative control and all cells with a greater binding signal, allowing us to enrich moderate-affinity binders. This process was repeated three times on the same day for a total of 300,000 cells. These cells were added to warm SOC+25 g/mL chloramphenicol and grown at 37° C. overnight. The next day, cells were frozen at −80° C. as glycerol stocks for MassTitr sorting.
MassTitr Sorting Scheme
The pre-enriched T7-Pep Library was induced and prepared as previously described. 7×107 cells were incubated with increasing concentrations of ENAH EVH1 tetramer (30, 12, 4.8, 1.9, 0.77, 0.31, 0.12, 0.049 μM) and then sorted on the BDFACS Aria. Cells were collected in 4 gates that were drawn with boundaries roughly parallel to the binding vs. expression signal slope for a series of positive controls (ActA, SHIP2, Vinculin). A no-peptide negative control (empty) was run to assess the degree of nonspecific binding at the time of sorting. At each concentration of ENAH EVH1 tetramer, cells were sorted for approximately 20 minutes and the number of cells that were collected in each gate was recorded. Enough cells were collected per gate to oversample the library at least 10-fold. Cells were collected in LB+0.2% w/v glucose and then sorted cells were transferred into 10 mL of LB+0.2% w/v glucose, grown at 37° C. overnight, and plasmid DNA was isolated by miniprep the next day. Three replicate Mass-Titr experiments were performed.
Illumina Amplicon Preparation
As described above, sorted pools were grown overnight at 37° C. in LB+0.2% glucose and then miniprepped (Qiagen). Samples collected at a certain gate per concentration of ENAH tetramer were each given a unique barcode/index combination. First, we PCR amplified the variable region of the library with a forward primer (Ngsfwd_1) and a corresponding reverse primer that contains one of 5 6-nucleotide (nt) index sequences for multiplexing (Ngsrev_1_i). 14 cycles of amplification were carried out with Phusion polymerase (NEB) using an annealing temperature of 66° C. The resulting reaction was PCR purified with the Zymo Clean and Concentrate kit (Zymo Research) and eluted with 20 μL milliQ water. 200 ng of each reaction was cut with MmeI (NEB) at 37° C. for 1 hour and then the enzyme was heat inactivated at 65° C. for 20 minutes. To the 5′ end of 15 μL of digested fragment we ligated a double stranded adapter with matching overhangs using T4 DNA ligase (NEB) at 25° C. for 30 minutes, followed by heat inactivation at 65° C. for 10 minutes. This adapter contains one of 24 5-nt barcodes and the standard Illumina forward primer sequence. The expected ˜200 bp band was gel purified with the Zymoclean Gel DNA recovery kit (Zymo Research) and eluted in 17 μL of MilliQ water. 3 μL of this reaction was PCR amplified for 10 cycles with primers Ngsfwd_2 and Ngsrev_2 in a 50 μL reaction at an annealing temperature of 66° C. The 5′ and 3′ Illumina adapter sequences and the reverse priming sequence were included in this step. The final product was PCR purified and eluted in MilliQ water. The DNA concentration of each pool was measured with the Qubit assay and samples were combined and run in one lane. In total, the multiplexed sample included 32 pools (corresponding to cells collected at each of 8 concentrations in 4 gates)×3 replicates, as well as the pre-enriched input library, for a total of 97 samples distinguished by their barcode/index combination. At each stage of the preparation process, the quality and homogeneity of the amplicon was assessed through Sanger sequencing (Genewiz) and Bioanalyzer. The multiplexed sample was submitted for sequencing on a NextSeq500 instrument using paired end reads.
MassTitr Data Processing
Sequences were demultiplexed and processed with an in-house script. We used the sequencing data to determine the number of clonal cells (i.e. cells displaying the same peptide) that were found in each gate at each concentration. We calculated the clone read frequency in gate j at concentration k as Rijk/Tijk, where Rijk is the number of raw reads for sequence i in gate j at concentration k, and Tjk is the total sequencing reads obtained for the pool corresponding to gate j at concentration k. This read frequency was multiplied by the total number of cells collected in each gate (recorded during the sorting) to calculate the total cell number Cijk in a given gate at each ENAH EVH1 tetramer concentration. This can be used to compute an effective gate position, given the gates collected, using equation 1:
where F1, F2, F3, and F4 are the mid y-axis fluorescence values for each gate (with values of 700, 1500, 2500, 4000 respectively) and citk is the total number of cells across all 4 gates for a given clone, obtained by summing cijk for j=1-4 at concentration k. Feffseq, con was calculated for each clone across all ENAH tetramer concentrations. If this experiment is performed in such a way such that every peptide-expressing cell is collected and sequenced, then the clones-per-gate information can be used to extract a concentration-dependent signal-vs-concentration curve that can be fit to give an apparent dissociation constant. In our experiment, we collected and sequenced fewer cells, focusing on cells that displayed above-background binding signal, and processed the data to obtain lower-resolution information about binding affinities.
Our data processing was focused on identifying a subset of well-behaved clones, of varying affinities, that we judged to be good candidates for subsequent analysis on binding features. We first removed clones for which we obtained <100 reads across all bins at all concentrations and also clones that did not have a total cell count (Cijk) of at least 25 cells in at least four concentrations. A sequence was assigned as an ENAH binder if (1) the clonal population of cells showed a concentration-dependent increase in binding signal in at least two of three replicate experiments, or (2) the displayed peptide contained an FP4 motif and showed concentration-dependent binding in at least one replicate. Among binders identified in this way, clones were further tagged as “high affinity” if they met additional requirements: (1) total cell count of greater than or equal to 80 cells for at least four concentrations, (2) greater than or equal to 10 cells found in binding gates at the lowest three concentrations (0.31 μM, 0.12 μM, 0.049 μM), (3) cells in more than one gate at the highest two concentrations (30 μM, 12 μM), and (4) clone found in at least two replicates. These filters are were chosen, by benchmarking against a few validated binding clones from the screen, to extract a subset of well-behaved clones that we judged likely to be true binders (as proved to be true).
MassTitr Sequence Analysis
The T7-pep library contains many point mutations, frameshifts, and prematurely terminated peptides, in addition to full-length human 36-mers. Consequently, for many hits we identified multiple closely related but non-identical sequences. To analyze trends in our dataset, we collapsed hits that we judged to be variants of the same sequence into one representative sequence, leading to a total of 108 non-redundant sequences. If one of the sequence variants matched exactly with a sequence from the proteome, that sequence was chosen as the representative sequence. Otherwise, the sequence with the most combined total cell counts was chosen for analysis. To compare these sequences to sequences from the pre-enriched input library, we clustered amino-acid sequences from the input library reads using CD-Hit with a sequence identity cut-off of 0.7, which effectively collapsed proteomic sequence variants into one representative sequence.
Identification of Putative Biological Interaction Partners and GO Analysis
To identify putative biological interaction partners among our MassTitr hits we first removed peptides that mapped to a human protein but contained an unnatural FP4 or CXC motif due to frameshift or point mutations. Then we identified those peptides predicted to be intrinsically disordered, using an IUPRED cutoff of >0.4. Finally, we assessed the cytoplasmic localization of fit hits using cellular component terms from QuickGO. Two proteins from our list (NHSL1 and KIAA1522) did not have any associated terms with our search criteria. For these proteins, we manually curated subcellular localizations from the literature and the Human Protein Atlas. For proteins reported to be membrane bound, such as MIA3, we confirmed that the regions we pulled out as hits were cytoplasmic as annotated in Uniprot. GO term enrichments were performed using PANTHER with a Fisher's Exact Test. GO biological process terms with an FDR<0.05 were designated as enriched.
Computational Rosetta Modelling
Input structure: An initial structure for the bidentate modeling pipeline was derived from
two separate, peptide bound EVH1-domain structures. Structure 5NC7 includes a short FP4-containing peptide (chain I) bound to the noncanonical site of the ENAH EVH1 domain (chain D), and the ENAH-ABI1 structure reported in this work provided a model of an FP8 peptide on the canonical site. These two structures were imported into PyMol and superimposed using the structural alignment function, and the FP8 peptide was truncated to be FP4. After verifying that the RMSD for the alignment was low (≤0.5 Å), the ENAH EVH1 domain of 5NC7 domain was deleted. The two structures were then merged and exported from PyMol. The resulting PDB contained an EVH1-domain bound to two independent FP4-containing peptides. The final structure was relaxed using the Rosetta Fast Relax protocol run under the default Rosetta energy function REF2015.
Chain Bridging: The Rosetta Bridge Chain mover, with the standard Rosetta energy function REF2015 with interchain centroid weights (interchain_cen), was used to link the noncanonical- and canonical-site-bound peptides of the seed structure. The insertion motif used for the protocol was “αLX”, which specified that the mover should generate backbone coordinates for a loop of length α composed of amino acids derived from any part of Ramachandran space. Starting with α=20, we tested shorter lengths of a until Bridge Chain could no longer find a solution. The FP4 motif seeds were anchored in their starting positions using distance constraints generated using the CoordinateConstraintGenerator mover in RosettaScripts with a strength/deviation parameter of 0.25 arbitrary units. The noncanonical-site peptide required a secondary set of constraints for stability, which bound each of its residues to within several angstroms of residue 45 of the EVH1 chain. The final structure, with the two motifs connected, was then relaxed using the Rosetta Fast Relax protocol run under the default Rosetta energy function REF2015. Because bidentate binding was possible in two different orientations while preserving the polarity of both FP4 motifs as found in the 5nc7 and ENAH-ABI1 structures, this process was done twice, once in each direction.
dTERMen
The dTERMen scoring function and protocol are described in Zhou, et al. (Proc Natl Acad Sci USA 2020 117:1059). The inputs to dTERMen are the backbone coordinates of a structure and a sequence. dTERMen returns a score for the input sequence adopting the input structure, with lower scores as better. Side-chain positions are not modeled explicitly. To score the EVL sequence on the ENAH-PCARE backbone template, we generated pairwise alignments of the EVH1 domains of ENAH and EVL to determine how to map sequence to structure. The EVL EVH1 domains is longer than that of ENAH, so unaligned residues were deleted. Specifically, Lys27 was removed from EVL. We then used dTERMen to score all possible combinations of residue swaps between EVL and ENAH, up to 6 possible positions. Residue swap combinations that led to the minimum energy score were recorded. The best 6 mutations were sufficient to nearly recapitulate the energy score of the native ENAH sequence on the ENAH PCARE template. We also included an I26A mutation based on manual inspect of the ENAH-PCARE828-848 structure. We cloned, over-expressed and purified this swapped EVL sequence, as described above, to test for binding to PCARE B
Plasmids for Cell Culture
For experiments in mammalian cells, the following plasmids were used: pLKO.1-Enah shRNA (GE Dharmacon TRCN0000061827, Homo sapiens antisense 5′-TTAGAGGAGTCTCAACAGAGG-3′), pLKO.1-Evl shRNA (Sigma TRCN0000091075, Mus musculus antisense 5′-TTGTTCATTTCTTCCATGAGG-3′), and non-targeting pLKO.1 control (a gift from Felicia Goodrum, University of Arizona). GFP, and mouse cDNA sequences for GFP-tagged ENAH, VASP, and EVL (gifts from Frank Gertler, MIT) were sub-cloned into the pCIB lentiviral expression vector (Addgene plasmid #120862) as previously described (Puleo et al. 2019). ENAH EVH1 domain deletion mutant was generated using inverse PCR site-directed mutagenesis of full-length ENAH. All sequences of constructed plasmids were confirmed by Sanger sequencing. mRuby2-Pcare and Mito-mRuby2-Pcare inserts were synthesized (Twist Bioscience) and subcloned into a SFFV-promoter lentiviral expression vector.
It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.
This application claims priority to U.S. Provisional Application No. 63/141,395, filed on Jan. 25, 2021. The disclosure of the prior application is considered part of and is incorporated by reference in the disclosure of this application.
The work described herein was conducted, at least in part, using funds from a federal grant. The government therefore has certain rights in the invention or inventions described in this specification.
| Number | Date | Country | |
|---|---|---|---|
| 63141395 | Jan 2021 | US |
